Pupils in grades K-6 will be trained in self-administration of verbal rewards to determine the effect of such training on self-concept. In classroom experimenters will be trained to use an observation and recording of behavior techniques which emphasizes responses of specific children to the self-administration of verbal rewards. Approximately 60 teachers will serve as experimenters. They will primarily work with a sample of children with negative self-concepts. The effects of this training on change in self-concept, locus of control, anxiety, and achievement will be observed. The sample will also be individually tested for verbal self-rewarding behavior and for self-rating of performance. The classroom experimenters will be observed using the Self-Concept Observer Rating scale once each month on a random time basis. Each observation will be for 15 minutes. The scores from these observations will examine the relationship between the degree of experimenter application of the treatments to the outcome measures. This study is a continuation of testing the modeling and reinforcement treatments in a natural school setting. The study views positive self-concept as, in part, the ability to give verbal self-reinforcement.